CIVICS: Building a Dataset for Examining Culturally-Informed Values in Large Language Models Paper • 2405.13974 • Published May 22, 2024 • 9
OpenCulture Collection A multilingual dataset of public domain books and newspapers. • 27 items • Updated Nov 6, 2024 • 121
KTO: Model Alignment as Prospect Theoretic Optimization Paper • 2402.01306 • Published Feb 2, 2024 • 16